1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 9 Bodo Bauer <bb@ricochet.net> 102.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 11move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 12fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 13===================== ======================================= ================ 14 15 16 17.. Table of Contents 18 19 0 Preface 20 0.1 Introduction/Credits 21 0.2 Legal Stuff 22 23 1 Collecting System Information 24 1.1 Process-Specific Subdirectories 25 1.2 Kernel data 26 1.3 IDE devices in /proc/ide 27 1.4 Networking info in /proc/net 28 1.5 SCSI info 29 1.6 Parallel port info in /proc/parport 30 1.7 TTY info in /proc/tty 31 1.8 Miscellaneous kernel statistics in /proc/stat 32 1.9 Ext4 file system parameters 33 34 2 Modifying System Parameters 35 36 3 Per-Process Parameters 37 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 38 score 39 3.2 /proc/<pid>/oom_score - Display current oom-killer score 40 3.3 /proc/<pid>/io - Display the IO accounting fields 41 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings 42 3.5 /proc/<pid>/mountinfo - Information about mounts 43 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 44 3.7 /proc/<pid>/task/<tid>/children - Information about task children 45 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 46 3.9 /proc/<pid>/map_files - Information about memory mapped files 47 3.10 /proc/<pid>/timerslack_ns - Task timerslack value 48 3.11 /proc/<pid>/patch_state - Livepatch patch operation state 49 3.12 /proc/<pid>/arch_status - Task architecture specific information 50 3.13 /proc/<pid>/fd - List of symlinks to open files 51 3.14 /proc/<pid/ksm_stat - Information about the process's ksm status. 52 53 4 Configuring procfs 54 4.1 Mount options 55 56 5 Filesystem behavior 57 58Preface 59======= 60 610.1 Introduction/Credits 62------------------------ 63 64This documentation is part of a soon (or so we hope) to be released book on 65the SuSE Linux distribution. As there is no complete documentation for the 66/proc file system and we've used many freely available sources to write these 67chapters, it seems only fair to give the work back to the Linux community. 68This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 69afraid it's still far from complete, but we hope it will be useful. As far as 70we know, it is the first 'all-in-one' document about the /proc file system. It 71is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 72SPARC, AXP, etc., features, you probably won't find what you are looking for. 73It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 74additions and patches are welcome and will be added to this document if you 75mail them to Bodo. 76 77We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 78other people for help compiling this documentation. We'd also like to extend a 79special thank you to Andi Kleen for documentation, which we relied on heavily 80to create this document, as well as the additional information he provided. 81Thanks to everybody else who contributed source or docs to the Linux kernel 82and helped create a great piece of software... :) 83 84If you have any comments, corrections or additions, please don't hesitate to 85contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 86document. 87 88The latest version of this document is available online at 89https://www.kernel.org/doc/html/latest/filesystems/proc.html 90 91If the above direction does not works for you, you could try the kernel 92mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 93comandante@zaralinux.com. 94 950.2 Legal Stuff 96--------------- 97 98We don't guarantee the correctness of this document, and if you come to us 99complaining about how you screwed up your system because of incorrect 100documentation, we won't feel responsible... 101 102Chapter 1: Collecting System Information 103======================================== 104 105In This Chapter 106--------------- 107* Investigating the properties of the pseudo file system /proc and its 108 ability to provide information on the running Linux system 109* Examining /proc's structure 110* Uncovering various information about the kernel and the processes running 111 on the system 112 113------------------------------------------------------------------------------ 114 115The proc file system acts as an interface to internal data structures in the 116kernel. It can be used to obtain information about the system and to change 117certain kernel parameters at runtime (sysctl). 118 119First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 120show you how you can use /proc/sys to change settings. 121 1221.1 Process-Specific Subdirectories 123----------------------------------- 124 125The directory /proc contains (among other things) one subdirectory for each 126process running on the system, which is named after the process ID (PID). 127 128The link 'self' points to the process reading the file system. Each process 129subdirectory has the entries listed in Table 1-1. 130 131A process can read its own information from /proc/PID/* with no extra 132permissions. When reading /proc/PID/* information for other processes, reading 133process is required to have either CAP_SYS_PTRACE capability with 134PTRACE_MODE_READ access permissions, or, alternatively, CAP_PERFMON 135capability. This applies to all read-only information like `maps`, `environ`, 136`pagemap`, etc. The only exception is `mem` file due to its read-write nature, 137which requires CAP_SYS_PTRACE capabilities with more elevated 138PTRACE_MODE_ATTACH permissions; CAP_PERFMON capability does not grant access 139to /proc/PID/mem for other processes. 140 141Note that an open file descriptor to /proc/<pid> or to any of its 142contained files or subdirectories does not prevent <pid> being reused 143for some other process in the event that <pid> exits. Operations on 144open /proc/<pid> file descriptors corresponding to dead processes 145never act on any new process that the kernel may, through chance, have 146also assigned the process ID <pid>. Instead, operations on these FDs 147usually fail with ESRCH. 148 149.. table:: Table 1-1: Process specific entries in /proc 150 151 ============= =============================================================== 152 File Content 153 ============= =============================================================== 154 clear_refs Clears page referenced bits shown in smaps output 155 cmdline Command line arguments 156 cpu Current and last cpu in which it was executed (2.4)(smp) 157 cwd Link to the current working directory 158 environ Values of environment variables 159 exe Link to the executable of this process 160 fd Directory, which contains all file descriptors 161 maps Memory maps to executables and library files (2.4) 162 mem Memory held by this process 163 root Link to the root directory of this process 164 stat Process status 165 statm Process memory status information 166 status Process status in human readable form 167 wchan Present with CONFIG_KALLSYMS=y: it shows the kernel function 168 symbol the task is blocked in - or "0" if not blocked. 169 pagemap Page table 170 stack Report full stack trace, enable via CONFIG_STACKTRACE 171 smaps An extension based on maps, showing the memory consumption of 172 each mapping and flags associated with it 173 smaps_rollup Accumulated smaps stats for all mappings of the process. This 174 can be derived from smaps, but is faster and more convenient 175 numa_maps An extension based on maps, showing the memory locality and 176 binding policy as well as mem usage (in pages) of each mapping. 177 ============= =============================================================== 178 179For example, to get the status information of a process, all you have to do is 180read the file /proc/PID/status:: 181 182 >cat /proc/self/status 183 Name: cat 184 State: R (running) 185 Tgid: 5452 186 Pid: 5452 187 PPid: 743 188 TracerPid: 0 (2.4) 189 Uid: 501 501 501 501 190 Gid: 100 100 100 100 191 FDSize: 256 192 Groups: 100 14 16 193 Kthread: 0 194 VmPeak: 5004 kB 195 VmSize: 5004 kB 196 VmLck: 0 kB 197 VmHWM: 476 kB 198 VmRSS: 476 kB 199 RssAnon: 352 kB 200 RssFile: 120 kB 201 RssShmem: 4 kB 202 VmData: 156 kB 203 VmStk: 88 kB 204 VmExe: 68 kB 205 VmLib: 1412 kB 206 VmPTE: 20 kb 207 VmSwap: 0 kB 208 HugetlbPages: 0 kB 209 CoreDumping: 0 210 THP_enabled: 1 211 Threads: 1 212 SigQ: 0/28578 213 SigPnd: 0000000000000000 214 ShdPnd: 0000000000000000 215 SigBlk: 0000000000000000 216 SigIgn: 0000000000000000 217 SigCgt: 0000000000000000 218 CapInh: 00000000fffffeff 219 CapPrm: 0000000000000000 220 CapEff: 0000000000000000 221 CapBnd: ffffffffffffffff 222 CapAmb: 0000000000000000 223 NoNewPrivs: 0 224 Seccomp: 0 225 Speculation_Store_Bypass: thread vulnerable 226 SpeculationIndirectBranch: conditional enabled 227 voluntary_ctxt_switches: 0 228 nonvoluntary_ctxt_switches: 1 229 230This shows you nearly the same information you would get if you viewed it with 231the ps command. In fact, ps uses the proc file system to obtain its 232information. But you get a more detailed view of the process by reading the 233file /proc/PID/status. It fields are described in table 1-2. 234 235The statm file contains more detailed information about the process 236memory usage. Its seven fields are explained in Table 1-3. The stat file 237contains detailed information about the process itself. Its fields are 238explained in Table 1-4. 239 240(for SMP CONFIG users) 241 242For making accounting scalable, RSS related information are handled in an 243asynchronous manner and the value may not be very precise. To see a precise 244snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 245It's slow but very precise. 246 247.. table:: Table 1-2: Contents of the status fields (as of 4.19) 248 249 ========================== =================================================== 250 Field Content 251 ========================== =================================================== 252 Name filename of the executable 253 Umask file mode creation mask 254 State state (R is running, S is sleeping, D is sleeping 255 in an uninterruptible wait, Z is zombie, 256 T is traced or stopped) 257 Tgid thread group ID 258 Ngid NUMA group ID (0 if none) 259 Pid process id 260 PPid process id of the parent process 261 TracerPid PID of process tracing this process (0 if not, or 262 the tracer is outside of the current pid namespace) 263 Uid Real, effective, saved set, and file system UIDs 264 Gid Real, effective, saved set, and file system GIDs 265 FDSize number of file descriptor slots currently allocated 266 Groups supplementary group list 267 NStgid descendant namespace thread group ID hierarchy 268 NSpid descendant namespace process ID hierarchy 269 NSpgid descendant namespace process group ID hierarchy 270 NSsid descendant namespace session ID hierarchy 271 Kthread kernel thread flag, 1 is yes, 0 is no 272 VmPeak peak virtual memory size 273 VmSize total program size 274 VmLck locked memory size 275 VmPin pinned memory size 276 VmHWM peak resident set size ("high water mark") 277 VmRSS size of memory portions. It contains the three 278 following parts 279 (VmRSS = RssAnon + RssFile + RssShmem) 280 RssAnon size of resident anonymous memory 281 RssFile size of resident file mappings 282 RssShmem size of resident shmem memory (includes SysV shm, 283 mapping of tmpfs and shared anonymous mappings) 284 VmData size of private data segments 285 VmStk size of stack segments 286 VmExe size of text segment 287 VmLib size of shared library code 288 VmPTE size of page table entries 289 VmSwap amount of swap used by anonymous private data 290 (shmem swap usage is not included) 291 HugetlbPages size of hugetlb memory portions 292 CoreDumping process's memory is currently being dumped 293 (killing the process may lead to a corrupted core) 294 THP_enabled process is allowed to use THP (returns 0 when 295 PR_SET_THP_DISABLE is set on the process 296 Threads number of threads 297 SigQ number of signals queued/max. number for queue 298 SigPnd bitmap of pending signals for the thread 299 ShdPnd bitmap of shared pending signals for the process 300 SigBlk bitmap of blocked signals 301 SigIgn bitmap of ignored signals 302 SigCgt bitmap of caught signals 303 CapInh bitmap of inheritable capabilities 304 CapPrm bitmap of permitted capabilities 305 CapEff bitmap of effective capabilities 306 CapBnd bitmap of capabilities bounding set 307 CapAmb bitmap of ambient capabilities 308 NoNewPrivs no_new_privs, like prctl(PR_GET_NO_NEW_PRIV, ...) 309 Seccomp seccomp mode, like prctl(PR_GET_SECCOMP, ...) 310 Speculation_Store_Bypass speculative store bypass mitigation status 311 SpeculationIndirectBranch indirect branch speculation mode 312 Cpus_allowed mask of CPUs on which this process may run 313 Cpus_allowed_list Same as previous, but in "list format" 314 Mems_allowed mask of memory nodes allowed to this process 315 Mems_allowed_list Same as previous, but in "list format" 316 voluntary_ctxt_switches number of voluntary context switches 317 nonvoluntary_ctxt_switches number of non voluntary context switches 318 ========================== =================================================== 319 320 321.. table:: Table 1-3: Contents of the statm fields (as of 2.6.8-rc3) 322 323 ======== =============================== ============================== 324 Field Content 325 ======== =============================== ============================== 326 size total program size (pages) (same as VmSize in status) 327 resident size of memory portions (pages) (same as VmRSS in status) 328 shared number of pages that are shared (i.e. backed by a file, same 329 as RssFile+RssShmem in status) 330 trs number of pages that are 'code' (not including libs; broken, 331 includes data segment) 332 lrs number of pages of library (always 0 on 2.6) 333 drs number of pages of data/stack (including libs; broken, 334 includes library text) 335 dt number of dirty pages (always 0 on 2.6) 336 ======== =============================== ============================== 337 338 339.. table:: Table 1-4: Contents of the stat fields (as of 2.6.30-rc7) 340 341 ============= =============================================================== 342 Field Content 343 ============= =============================================================== 344 pid process id 345 tcomm filename of the executable 346 state state (R is running, S is sleeping, D is sleeping in an 347 uninterruptible wait, Z is zombie, T is traced or stopped) 348 ppid process id of the parent process 349 pgrp pgrp of the process 350 sid session id 351 tty_nr tty the process uses 352 tty_pgrp pgrp of the tty 353 flags task flags 354 min_flt number of minor faults 355 cmin_flt number of minor faults with child's 356 maj_flt number of major faults 357 cmaj_flt number of major faults with child's 358 utime user mode jiffies 359 stime kernel mode jiffies 360 cutime user mode jiffies with child's 361 cstime kernel mode jiffies with child's 362 priority priority level 363 nice nice level 364 num_threads number of threads 365 it_real_value (obsolete, always 0) 366 start_time time the process started after system boot 367 vsize virtual memory size 368 rss resident set memory size 369 rsslim current limit in bytes on the rss 370 start_code address above which program text can run 371 end_code address below which program text can run 372 start_stack address of the start of the main process stack 373 esp current value of ESP 374 eip current value of EIP 375 pending bitmap of pending signals 376 blocked bitmap of blocked signals 377 sigign bitmap of ignored signals 378 sigcatch bitmap of caught signals 379 0 (place holder, used to be the wchan address, 380 use /proc/PID/wchan instead) 381 0 (place holder) 382 0 (place holder) 383 exit_signal signal to send to parent thread on exit 384 task_cpu which CPU the task is scheduled on 385 rt_priority realtime priority 386 policy scheduling policy (man sched_setscheduler) 387 blkio_ticks time spent waiting for block IO 388 gtime guest time of the task in jiffies 389 cgtime guest time of the task children in jiffies 390 start_data address above which program data+bss is placed 391 end_data address below which program data+bss is placed 392 start_brk address above which program heap can be expanded with brk() 393 arg_start address above which program command line is placed 394 arg_end address below which program command line is placed 395 env_start address above which program environment is placed 396 env_end address below which program environment is placed 397 exit_code the thread's exit_code in the form reported by the waitpid 398 system call 399 ============= =============================================================== 400 401The /proc/PID/maps file contains the currently mapped memory regions and 402their access permissions. 403 404The format is:: 405 406 address perms offset dev inode pathname 407 408 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 409 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 410 0804a000-0806b000 rw-p 00000000 00:00 0 [heap] 411 a7cb1000-a7cb2000 ---p 00000000 00:00 0 412 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 413 a7eb2000-a7eb3000 ---p 00000000 00:00 0 414 a7eb3000-a7ed5000 rw-p 00000000 00:00 0 415 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 416 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 417 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 418 a800b000-a800e000 rw-p 00000000 00:00 0 419 a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 420 a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 421 a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 422 a8024000-a8027000 rw-p 00000000 00:00 0 423 a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 424 a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 425 a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 426 aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 427 ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 428 429where "address" is the address space in the process that it occupies, "perms" 430is a set of permissions:: 431 432 r = read 433 w = write 434 x = execute 435 s = shared 436 p = private (copy on write) 437 438"offset" is the offset into the mapping, "dev" is the device (major:minor), and 439"inode" is the inode on that device. 0 indicates that no inode is associated 440with the memory region, as the case would be with BSS (uninitialized data). 441The "pathname" shows the name associated file for this mapping. If the mapping 442is not associated with a file: 443 444 =================== =========================================== 445 [heap] the heap of the program 446 [stack] the stack of the main process 447 [vdso] the "virtual dynamic shared object", 448 the kernel system call handler 449 [anon:<name>] a private anonymous mapping that has been 450 named by userspace 451 [anon_shmem:<name>] an anonymous shared memory mapping that has 452 been named by userspace 453 =================== =========================================== 454 455 or if empty, the mapping is anonymous. 456 457Starting with 6.11 kernel, /proc/PID/maps provides an alternative 458ioctl()-based API that gives ability to flexibly and efficiently query and 459filter individual VMAs. This interface is binary and is meant for more 460efficient and easy programmatic use. `struct procmap_query`, defined in 461linux/fs.h UAPI header, serves as an input/output argument to the 462`PROCMAP_QUERY` ioctl() command. See comments in linus/fs.h UAPI header for 463details on query semantics, supported flags, data returned, and general API 464usage information. 465 466The /proc/PID/smaps is an extension based on maps, showing the memory 467consumption for each of the process's mappings. For each mapping (aka Virtual 468Memory Area, or VMA) there is a series of lines such as the following:: 469 470 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 471 472 Size: 1084 kB 473 KernelPageSize: 4 kB 474 MMUPageSize: 4 kB 475 Rss: 892 kB 476 Pss: 374 kB 477 Pss_Dirty: 0 kB 478 Shared_Clean: 892 kB 479 Shared_Dirty: 0 kB 480 Private_Clean: 0 kB 481 Private_Dirty: 0 kB 482 Referenced: 892 kB 483 Anonymous: 0 kB 484 KSM: 0 kB 485 LazyFree: 0 kB 486 AnonHugePages: 0 kB 487 ShmemPmdMapped: 0 kB 488 Shared_Hugetlb: 0 kB 489 Private_Hugetlb: 0 kB 490 Swap: 0 kB 491 SwapPss: 0 kB 492 KernelPageSize: 4 kB 493 MMUPageSize: 4 kB 494 Locked: 0 kB 495 THPeligible: 0 496 VmFlags: rd ex mr mw me dw 497 498The first of these lines shows the same information as is displayed for 499the mapping in /proc/PID/maps. Following lines show the size of the 500mapping (size); the size of each page allocated when backing a VMA 501(KernelPageSize), which is usually the same as the size in the page table 502entries; the page size used by the MMU when backing a VMA (in most cases, 503the same as KernelPageSize); the amount of the mapping that is currently 504resident in RAM (RSS); the process's proportional share of this mapping 505(PSS); and the number of clean and dirty shared and private pages in the 506mapping. 507 508The "proportional set size" (PSS) of a process is the count of pages it has 509in memory, where each page is divided by the number of processes sharing it. 510So if a process has 1000 pages all to itself, and 1000 shared with one other 511process, its PSS will be 1500. "Pss_Dirty" is the portion of PSS which 512consists of dirty pages. ("Pss_Clean" is not included, but it can be 513calculated by subtracting "Pss_Dirty" from "Pss".) 514 515Note that even a page which is part of a MAP_SHARED mapping, but has only 516a single pte mapped, i.e. is currently used by only one process, is accounted 517as private and not as shared. 518 519"Referenced" indicates the amount of memory currently marked as referenced or 520accessed. 521 522"Anonymous" shows the amount of memory that does not belong to any file. Even 523a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 524and a page is modified, the file page is replaced by a private anonymous copy. 525 526"KSM" reports how many of the pages are KSM pages. Note that KSM-placed zeropages 527are not included, only actual KSM pages. 528 529"LazyFree" shows the amount of memory which is marked by madvise(MADV_FREE). 530The memory isn't freed immediately with madvise(). It's freed in memory 531pressure if the memory is clean. Please note that the printed value might 532be lower than the real value due to optimizations used in the current 533implementation. If this is not desirable please file a bug report. 534 535"AnonHugePages" shows the amount of memory backed by transparent hugepage. 536 537"ShmemPmdMapped" shows the amount of shared (shmem/tmpfs) memory backed by 538huge pages. 539 540"Shared_Hugetlb" and "Private_Hugetlb" show the amounts of memory backed by 541hugetlbfs page which is *not* counted in "RSS" or "PSS" field for historical 542reasons. And these are not included in {Shared,Private}_{Clean,Dirty} field. 543 544"Swap" shows how much would-be-anonymous memory is also used, but out on swap. 545 546For shmem mappings, "Swap" includes also the size of the mapped (and not 547replaced by copy-on-write) part of the underlying shmem object out on swap. 548"SwapPss" shows proportional swap share of this mapping. Unlike "Swap", this 549does not take into account swapped out page of underlying shmem objects. 550"Locked" indicates whether the mapping is locked in memory or not. 551 552"THPeligible" indicates whether the mapping is eligible for allocating 553naturally aligned THP pages of any currently enabled size. 1 if true, 0 554otherwise. 555 556"VmFlags" field deserves a separate description. This member represents the 557kernel flags associated with the particular virtual memory area in two letter 558encoded manner. The codes are the following: 559 560 == ======================================= 561 rd readable 562 wr writeable 563 ex executable 564 sh shared 565 mr may read 566 mw may write 567 me may execute 568 ms may share 569 gd stack segment growns down 570 pf pure PFN range 571 dw disabled write to the mapped file 572 lo pages are locked in memory 573 io memory mapped I/O area 574 sr sequential read advise provided 575 rr random read advise provided 576 dc do not copy area on fork 577 de do not expand area on remapping 578 ac area is accountable 579 nr swap space is not reserved for the area 580 ht area uses huge tlb pages 581 sf synchronous page fault 582 ar architecture specific flag 583 wf wipe on fork 584 dd do not include area into core dump 585 sd soft dirty flag 586 mm mixed map area 587 hg huge page advise flag 588 nh no huge page advise flag 589 mg mergeable advise flag 590 bt arm64 BTI guarded page 591 mt arm64 MTE allocation tags are enabled 592 um userfaultfd missing tracking 593 uw userfaultfd wr-protect tracking 594 ss shadow/guarded control stack page 595 sl sealed 596 == ======================================= 597 598Note that there is no guarantee that every flag and associated mnemonic will 599be present in all further kernel releases. Things get changed, the flags may 600be vanished or the reverse -- new added. Interpretation of their meaning 601might change in future as well. So each consumer of these flags has to 602follow each specific kernel version for the exact semantic. 603 604This file is only present if the CONFIG_MMU kernel configuration option is 605enabled. 606 607Note: reading /proc/PID/maps or /proc/PID/smaps is inherently racy (consistent 608output can be achieved only in the single read call). 609 610This typically manifests when doing partial reads of these files while the 611memory map is being modified. Despite the races, we do provide the following 612guarantees: 613 6141) The mapped addresses never go backwards, which implies no two 615 regions will ever overlap. 6162) If there is something at a given vaddr during the entirety of the 617 life of the smaps/maps walk, there will be some output for it. 618 619The /proc/PID/smaps_rollup file includes the same fields as /proc/PID/smaps, 620but their values are the sums of the corresponding values for all mappings of 621the process. Additionally, it contains these fields: 622 623- Pss_Anon 624- Pss_File 625- Pss_Shmem 626 627They represent the proportional shares of anonymous, file, and shmem pages, as 628described for smaps above. These fields are omitted in smaps since each 629mapping identifies the type (anon, file, or shmem) of all pages it contains. 630Thus all information in smaps_rollup can be derived from smaps, but at a 631significantly higher cost. 632 633The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 634bits on both physical and virtual pages associated with a process, and the 635soft-dirty bit on pte (see Documentation/admin-guide/mm/soft-dirty.rst 636for details). 637To clear the bits for all the pages associated with the process:: 638 639 > echo 1 > /proc/PID/clear_refs 640 641To clear the bits for the anonymous pages associated with the process:: 642 643 > echo 2 > /proc/PID/clear_refs 644 645To clear the bits for the file mapped pages associated with the process:: 646 647 > echo 3 > /proc/PID/clear_refs 648 649To clear the soft-dirty bit:: 650 651 > echo 4 > /proc/PID/clear_refs 652 653To reset the peak resident set size ("high water mark") to the process's 654current value:: 655 656 > echo 5 > /proc/PID/clear_refs 657 658Any other value written to /proc/PID/clear_refs will have no effect. 659 660The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 661using /proc/kpageflags and number of times a page is mapped using 662/proc/kpagecount. For detailed explanation, see 663Documentation/admin-guide/mm/pagemap.rst. 664 665The /proc/pid/numa_maps is an extension based on maps, showing the memory 666locality and binding policy, as well as the memory usage (in pages) of 667each mapping. The output follows a general format where mapping details get 668summarized separated by blank spaces, one mapping per each file line:: 669 670 address policy mapping details 671 672 00400000 default file=/usr/local/bin/app mapped=1 active=0 N3=1 kernelpagesize_kB=4 673 00600000 default file=/usr/local/bin/app anon=1 dirty=1 N3=1 kernelpagesize_kB=4 674 3206000000 default file=/lib64/ld-2.12.so mapped=26 mapmax=6 N0=24 N3=2 kernelpagesize_kB=4 675 320621f000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 676 3206220000 default file=/lib64/ld-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 677 3206221000 default anon=1 dirty=1 N3=1 kernelpagesize_kB=4 678 3206800000 default file=/lib64/libc-2.12.so mapped=59 mapmax=21 active=55 N0=41 N3=18 kernelpagesize_kB=4 679 320698b000 default file=/lib64/libc-2.12.so 680 3206b8a000 default file=/lib64/libc-2.12.so anon=2 dirty=2 N3=2 kernelpagesize_kB=4 681 3206b8e000 default file=/lib64/libc-2.12.so anon=1 dirty=1 N3=1 kernelpagesize_kB=4 682 3206b8f000 default anon=3 dirty=3 active=1 N3=3 kernelpagesize_kB=4 683 7f4dc10a2000 default anon=3 dirty=3 N3=3 kernelpagesize_kB=4 684 7f4dc10b4000 default anon=2 dirty=2 active=1 N3=2 kernelpagesize_kB=4 685 7f4dc1200000 default file=/anon_hugepage\040(deleted) huge anon=1 dirty=1 N3=1 kernelpagesize_kB=2048 686 7fff335f0000 default stack anon=3 dirty=3 N3=3 kernelpagesize_kB=4 687 7fff3369d000 default mapped=1 mapmax=35 active=0 N3=1 kernelpagesize_kB=4 688 689Where: 690 691"address" is the starting address for the mapping; 692 693"policy" reports the NUMA memory policy set for the mapping (see Documentation/admin-guide/mm/numa_memory_policy.rst); 694 695"mapping details" summarizes mapping data such as mapping type, page usage counters, 696node locality page counters (N0 == node0, N1 == node1, ...) and the kernel page 697size, in KB, that is backing the mapping up. 698 6991.2 Kernel data 700--------------- 701 702Similar to the process entries, the kernel data files give information about 703the running kernel. The files used to obtain this information are contained in 704/proc and are listed in Table 1-5. Not all of these will be present in your 705system. It depends on the kernel configuration and the loaded modules, which 706files are there, and which are missing. 707 708.. table:: Table 1-5: Kernel info in /proc 709 710 ============ =============================================================== 711 File Content 712 ============ =============================================================== 713 allocinfo Memory allocations profiling information 714 apm Advanced power management info 715 bootconfig Kernel command line obtained from boot config, 716 and, if there were kernel parameters from the 717 boot loader, a "# Parameters from bootloader:" 718 line followed by a line containing those 719 parameters prefixed by "# ". (5.5) 720 buddyinfo Kernel memory allocator information (see text) (2.5) 721 bus Directory containing bus specific information 722 cmdline Kernel command line, both from bootloader and embedded 723 in the kernel image 724 cpuinfo Info about the CPU 725 devices Available devices (block and character) 726 dma Used DMS channels 727 filesystems Supported filesystems 728 driver Various drivers grouped here, currently rtc (2.4) 729 execdomains Execdomains, related to security (2.4) 730 fb Frame Buffer devices (2.4) 731 fs File system parameters, currently nfs/exports (2.4) 732 ide Directory containing info about the IDE subsystem 733 interrupts Interrupt usage 734 iomem Memory map (2.4) 735 ioports I/O port usage 736 irq Masks for irq to cpu affinity (2.4)(smp?) 737 isapnp ISA PnP (Plug&Play) Info (2.4) 738 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 739 kmsg Kernel messages 740 ksyms Kernel symbol table 741 loadavg Load average of last 1, 5 & 15 minutes; 742 number of processes currently runnable (running or on ready queue); 743 total number of processes in system; 744 last pid created. 745 All fields are separated by one space except "number of 746 processes currently runnable" and "total number of processes 747 in system", which are separated by a slash ('/'). Example: 748 0.61 0.61 0.55 3/828 22084 749 locks Kernel locks 750 meminfo Memory info 751 misc Miscellaneous 752 modules List of loaded modules 753 mounts Mounted filesystems 754 net Networking info (see text) 755 pagetypeinfo Additional page allocator information (see text) (2.5) 756 partitions Table of partitions known to the system 757 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 758 decoupled by lspci (2.4) 759 rtc Real time clock 760 scsi SCSI info (see text) 761 slabinfo Slab pool info 762 softirqs softirq usage 763 stat Overall statistics 764 swaps Swap space utilization 765 sys See chapter 2 766 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 767 tty Info of tty drivers 768 uptime Wall clock since boot, combined idle time of all cpus 769 version Kernel version 770 video bttv info of video resources (2.4) 771 vmallocinfo Show vmalloced areas 772 ============ =============================================================== 773 774You can, for example, check which interrupts are currently in use and what 775they are used for by looking in the file /proc/interrupts:: 776 777 > cat /proc/interrupts 778 CPU0 779 0: 8728810 XT-PIC timer 780 1: 895 XT-PIC keyboard 781 2: 0 XT-PIC cascade 782 3: 531695 XT-PIC aha152x 783 4: 2014133 XT-PIC serial 784 5: 44401 XT-PIC pcnet_cs 785 8: 2 XT-PIC rtc 786 11: 8 XT-PIC i82365 787 12: 182918 XT-PIC PS/2 Mouse 788 13: 1 XT-PIC fpu 789 14: 1232265 XT-PIC ide0 790 15: 7 XT-PIC ide1 791 NMI: 0 792 793In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 794output of a SMP machine):: 795 796 > cat /proc/interrupts 797 798 CPU0 CPU1 799 0: 1243498 1214548 IO-APIC-edge timer 800 1: 8949 8958 IO-APIC-edge keyboard 801 2: 0 0 XT-PIC cascade 802 5: 11286 10161 IO-APIC-edge soundblaster 803 8: 1 0 IO-APIC-edge rtc 804 9: 27422 27407 IO-APIC-edge 3c503 805 12: 113645 113873 IO-APIC-edge PS/2 Mouse 806 13: 0 0 XT-PIC fpu 807 14: 22491 24012 IO-APIC-edge ide0 808 15: 2183 2415 IO-APIC-edge ide1 809 17: 30564 30414 IO-APIC-level eth0 810 18: 177 164 IO-APIC-level bttv 811 NMI: 2457961 2457959 812 LOC: 2457882 2457881 813 ERR: 2155 814 815NMI is incremented in this case because every timer interrupt generates a NMI 816(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 817 818LOC is the local interrupt counter of the internal APIC of every CPU. 819 820ERR is incremented in the case of errors in the IO-APIC bus (the bus that 821connects the CPUs in a SMP system. This means that an error has been detected, 822the IO-APIC automatically retry the transmission, so it should not be a big 823problem, but you should read the SMP-FAQ. 824 825In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 826/proc/interrupts to display every IRQ vector in use by the system, not 827just those considered 'most important'. The new vectors are: 828 829THR 830 interrupt raised when a machine check threshold counter 831 (typically counting ECC corrected errors of memory or cache) exceeds 832 a configurable threshold. Only available on some systems. 833 834TRM 835 a thermal event interrupt occurs when a temperature threshold 836 has been exceeded for the CPU. This interrupt may also be generated 837 when the temperature drops back to normal. 838 839SPU 840 a spurious interrupt is some interrupt that was raised then lowered 841 by some IO device before it could be fully processed by the APIC. Hence 842 the APIC sees the interrupt but does not know what device it came from. 843 For this case the APIC will generate the interrupt with a IRQ vector 844 of 0xff. This might also be generated by chipset bugs. 845 846RES, CAL, TLB 847 rescheduling, call and TLB flush interrupts are 848 sent from one CPU to another per the needs of the OS. Typically, 849 their statistics are used by kernel developers and interested users to 850 determine the occurrence of interrupts of the given type. 851 852The above IRQ vectors are displayed only when relevant. For example, 853the threshold vector does not exist on x86_64 platforms. Others are 854suppressed when the system is a uniprocessor. As of this writing, only 855i386 and x86_64 platforms support the new IRQ vector displays. 856 857Of some interest is the introduction of the /proc/irq directory to 2.4. 858It could be used to set IRQ to CPU affinity. This means that you can "hook" an 859IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 860irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 861prof_cpu_mask. 862 863For example:: 864 865 > ls /proc/irq/ 866 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 867 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 868 > ls /proc/irq/0/ 869 smp_affinity 870 871smp_affinity is a bitmask, in which you can specify which CPUs can handle the 872IRQ. You can set it by doing:: 873 874 > echo 1 > /proc/irq/10/smp_affinity 875 876This means that only the first CPU will handle the IRQ, but you can also echo 8775 which means that only the first and third CPU can handle the IRQ. 878 879The contents of each smp_affinity file is the same by default:: 880 881 > cat /proc/irq/0/smp_affinity 882 ffffffff 883 884There is an alternate interface, smp_affinity_list which allows specifying 885a CPU range instead of a bitmask:: 886 887 > cat /proc/irq/0/smp_affinity_list 888 1024-1031 889 890The default_smp_affinity mask applies to all non-active IRQs, which are the 891IRQs which have not yet been allocated/activated, and hence which lack a 892/proc/irq/[0-9]* directory. 893 894The node file on an SMP system shows the node to which the device using the IRQ 895reports itself as being attached. This hardware locality information does not 896include information about any possible driver locality preference. 897 898prof_cpu_mask specifies which CPUs are to be profiled by the system wide 899profiler. Default value is ffffffff (all CPUs if there are only 32 of them). 900 901The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 902between all the CPUs which are allowed to handle it. As usual the kernel has 903more info than you and does a better job than you, so the defaults are the 904best choice for almost everyone. [Note this applies only to those IO-APIC's 905that support "Round Robin" interrupt distribution.] 906 907There are three more important subdirectories in /proc: net, scsi, and sys. 908The general rule is that the contents, or even the existence of these 909directories, depend on your kernel configuration. If SCSI is not enabled, the 910directory scsi may not exist. The same is true with the net, which is there 911only when networking support is present in the running kernel. 912 913The slabinfo file gives information about memory usage at the slab level. 914Linux uses slab pools for memory management above page level in version 2.2. 915Commonly used objects have their own slab pool (such as network buffers, 916directory cache, and so on). 917 918:: 919 920 > cat /proc/buddyinfo 921 922 Node 0, zone DMA 0 4 5 4 4 3 ... 923 Node 0, zone Normal 1 0 0 1 101 8 ... 924 Node 0, zone HighMem 2 0 0 1 1 0 ... 925 926External fragmentation is a problem under some workloads, and buddyinfo is a 927useful tool for helping diagnose these problems. Buddyinfo will give you a 928clue as to how big an area you can safely allocate, or why a previous 929allocation failed. 930 931Each column represents the number of pages of a certain order which are 932available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 933ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 934available in ZONE_NORMAL, etc... 935 936More information relevant to external fragmentation can be found in 937pagetypeinfo:: 938 939 > cat /proc/pagetypeinfo 940 Page block order: 9 941 Pages per block: 512 942 943 Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 944 Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 945 Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 946 Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 947 Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 948 Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 949 Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 950 Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 951 Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 952 Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 953 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 954 955 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 956 Node 0, zone DMA 2 0 5 1 0 957 Node 0, zone DMA32 41 6 967 2 0 958 959Fragmentation avoidance in the kernel works by grouping pages of different 960migrate types into the same contiguous regions of memory called page blocks. 961A page block is typically the size of the default hugepage size, e.g. 2MB on 962X86-64. By keeping pages grouped based on their ability to move, the kernel 963can reclaim pages within a page block to satisfy a high-order allocation. 964 965The pagetypinfo begins with information on the size of a page block. It 966then gives the same type of information as buddyinfo except broken down 967by migrate-type and finishes with details on how many page blocks of each 968type exist. 969 970If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 971from libhugetlbfs https://github.com/libhugetlbfs/libhugetlbfs/), one can 972make an estimate of the likely number of huge pages that can be allocated 973at a given point in time. All the "Movable" blocks should be allocatable 974unless memory has been mlock()'d. Some of the Reclaimable blocks should 975also be allocatable although a lot of filesystem metadata may have to be 976reclaimed to achieve this. 977 978 979allocinfo 980~~~~~~~~~ 981 982Provides information about memory allocations at all locations in the code 983base. Each allocation in the code is identified by its source file, line 984number, module (if originates from a loadable module) and the function calling 985the allocation. The number of bytes allocated and number of calls at each 986location are reported. The first line indicates the version of the file, the 987second line is the header listing fields in the file. 988 989Example output. 990 991:: 992 993 > tail -n +3 /proc/allocinfo | sort -rn 994 127664128 31168 mm/page_ext.c:270 func:alloc_page_ext 995 56373248 4737 mm/slub.c:2259 func:alloc_slab_page 996 14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded 997 14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash 998 13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs 999 11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio 1000 9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node 1001 4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable 1002 4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start 1003 3940352 962 mm/memory.c:4214 func:alloc_anon_folio 1004 2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node 1005 ... 1006 1007 1008meminfo 1009~~~~~~~ 1010 1011Provides information about distribution and utilization of memory. This 1012varies by architecture and compile options. Some of the counters reported 1013here overlap. The memory reported by the non overlapping counters may not 1014add up to the overall memory usage and the difference for some workloads 1015can be substantial. In many cases there are other means to find out 1016additional memory using subsystem specific interfaces, for instance 1017/proc/net/sockstat for TCP memory allocations. 1018 1019Example output. You may not have all of these fields. 1020 1021:: 1022 1023 > cat /proc/meminfo 1024 1025 MemTotal: 32858820 kB 1026 MemFree: 21001236 kB 1027 MemAvailable: 27214312 kB 1028 Buffers: 581092 kB 1029 Cached: 5587612 kB 1030 SwapCached: 0 kB 1031 Active: 3237152 kB 1032 Inactive: 7586256 kB 1033 Active(anon): 94064 kB 1034 Inactive(anon): 4570616 kB 1035 Active(file): 3143088 kB 1036 Inactive(file): 3015640 kB 1037 Unevictable: 0 kB 1038 Mlocked: 0 kB 1039 SwapTotal: 0 kB 1040 SwapFree: 0 kB 1041 Zswap: 1904 kB 1042 Zswapped: 7792 kB 1043 Dirty: 12 kB 1044 Writeback: 0 kB 1045 AnonPages: 4654780 kB 1046 Mapped: 266244 kB 1047 Shmem: 9976 kB 1048 KReclaimable: 517708 kB 1049 Slab: 660044 kB 1050 SReclaimable: 517708 kB 1051 SUnreclaim: 142336 kB 1052 KernelStack: 11168 kB 1053 PageTables: 20540 kB 1054 SecPageTables: 0 kB 1055 NFS_Unstable: 0 kB 1056 Bounce: 0 kB 1057 WritebackTmp: 0 kB 1058 CommitLimit: 16429408 kB 1059 Committed_AS: 7715148 kB 1060 VmallocTotal: 34359738367 kB 1061 VmallocUsed: 40444 kB 1062 VmallocChunk: 0 kB 1063 Percpu: 29312 kB 1064 EarlyMemtestBad: 0 kB 1065 HardwareCorrupted: 0 kB 1066 AnonHugePages: 4149248 kB 1067 ShmemHugePages: 0 kB 1068 ShmemPmdMapped: 0 kB 1069 FileHugePages: 0 kB 1070 FilePmdMapped: 0 kB 1071 CmaTotal: 0 kB 1072 CmaFree: 0 kB 1073 HugePages_Total: 0 1074 HugePages_Free: 0 1075 HugePages_Rsvd: 0 1076 HugePages_Surp: 0 1077 Hugepagesize: 2048 kB 1078 Hugetlb: 0 kB 1079 DirectMap4k: 401152 kB 1080 DirectMap2M: 10008576 kB 1081 DirectMap1G: 24117248 kB 1082 1083MemTotal 1084 Total usable RAM (i.e. physical RAM minus a few reserved 1085 bits and the kernel binary code) 1086MemFree 1087 Total free RAM. On highmem systems, the sum of LowFree+HighFree 1088MemAvailable 1089 An estimate of how much memory is available for starting new 1090 applications, without swapping. Calculated from MemFree, 1091 SReclaimable, the size of the file LRU lists, and the low 1092 watermarks in each zone. 1093 The estimate takes into account that the system needs some 1094 page cache to function well, and that not all reclaimable 1095 slab will be reclaimable, due to items being in use. The 1096 impact of those factors will vary from system to system. 1097Buffers 1098 Relatively temporary storage for raw disk blocks 1099 shouldn't get tremendously large (20MB or so) 1100Cached 1101 In-memory cache for files read from the disk (the 1102 pagecache) as well as tmpfs & shmem. 1103 Doesn't include SwapCached. 1104SwapCached 1105 Memory that once was swapped out, is swapped back in but 1106 still also is in the swapfile (if memory is needed it 1107 doesn't need to be swapped out AGAIN because it is already 1108 in the swapfile. This saves I/O) 1109Active 1110 Memory that has been used more recently and usually not 1111 reclaimed unless absolutely necessary. 1112Inactive 1113 Memory which has been less recently used. It is more 1114 eligible to be reclaimed for other purposes 1115Unevictable 1116 Memory allocated for userspace which cannot be reclaimed, such 1117 as mlocked pages, ramfs backing pages, secret memfd pages etc. 1118Mlocked 1119 Memory locked with mlock(). 1120HighTotal, HighFree 1121 Highmem is all memory above ~860MB of physical memory. 1122 Highmem areas are for use by userspace programs, or 1123 for the pagecache. The kernel must use tricks to access 1124 this memory, making it slower to access than lowmem. 1125LowTotal, LowFree 1126 Lowmem is memory which can be used for everything that 1127 highmem can be used for, but it is also available for the 1128 kernel's use for its own data structures. Among many 1129 other things, it is where everything from the Slab is 1130 allocated. Bad things happen when you're out of lowmem. 1131SwapTotal 1132 total amount of swap space available 1133SwapFree 1134 Memory which has been evicted from RAM, and is temporarily 1135 on the disk 1136Zswap 1137 Memory consumed by the zswap backend (compressed size) 1138Zswapped 1139 Amount of anonymous memory stored in zswap (original size) 1140Dirty 1141 Memory which is waiting to get written back to the disk 1142Writeback 1143 Memory which is actively being written back to the disk 1144AnonPages 1145 Non-file backed pages mapped into userspace page tables 1146Mapped 1147 files which have been mmapped, such as libraries 1148Shmem 1149 Total memory used by shared memory (shmem) and tmpfs 1150KReclaimable 1151 Kernel allocations that the kernel will attempt to reclaim 1152 under memory pressure. Includes SReclaimable (below), and other 1153 direct allocations with a shrinker. 1154Slab 1155 in-kernel data structures cache 1156SReclaimable 1157 Part of Slab, that might be reclaimed, such as caches 1158SUnreclaim 1159 Part of Slab, that cannot be reclaimed on memory pressure 1160KernelStack 1161 Memory consumed by the kernel stacks of all tasks 1162PageTables 1163 Memory consumed by userspace page tables 1164SecPageTables 1165 Memory consumed by secondary page tables, this currently includes 1166 KVM mmu and IOMMU allocations on x86 and arm64. 1167NFS_Unstable 1168 Always zero. Previous counted pages which had been written to 1169 the server, but has not been committed to stable storage. 1170Bounce 1171 Memory used for block device "bounce buffers" 1172WritebackTmp 1173 Memory used by FUSE for temporary writeback buffers 1174CommitLimit 1175 Based on the overcommit ratio ('vm.overcommit_ratio'), 1176 this is the total amount of memory currently available to 1177 be allocated on the system. This limit is only adhered to 1178 if strict overcommit accounting is enabled (mode 2 in 1179 'vm.overcommit_memory'). 1180 1181 The CommitLimit is calculated with the following formula:: 1182 1183 CommitLimit = ([total RAM pages] - [total huge TLB pages]) * 1184 overcommit_ratio / 100 + [total swap pages] 1185 1186 For example, on a system with 1G of physical RAM and 7G 1187 of swap with a `vm.overcommit_ratio` of 30 it would 1188 yield a CommitLimit of 7.3G. 1189 1190 For more details, see the memory overcommit documentation 1191 in mm/overcommit-accounting. 1192Committed_AS 1193 The amount of memory presently allocated on the system. 1194 The committed memory is a sum of all of the memory which 1195 has been allocated by processes, even if it has not been 1196 "used" by them as of yet. A process which malloc()'s 1G 1197 of memory, but only touches 300M of it will show up as 1198 using 1G. This 1G is memory which has been "committed" to 1199 by the VM and can be used at any time by the allocating 1200 application. With strict overcommit enabled on the system 1201 (mode 2 in 'vm.overcommit_memory'), allocations which would 1202 exceed the CommitLimit (detailed above) will not be permitted. 1203 This is useful if one needs to guarantee that processes will 1204 not fail due to lack of memory once that memory has been 1205 successfully allocated. 1206VmallocTotal 1207 total size of vmalloc virtual address space 1208VmallocUsed 1209 amount of vmalloc area which is used 1210VmallocChunk 1211 largest contiguous block of vmalloc area which is free 1212Percpu 1213 Memory allocated to the percpu allocator used to back percpu 1214 allocations. This stat excludes the cost of metadata. 1215EarlyMemtestBad 1216 The amount of RAM/memory in kB, that was identified as corrupted 1217 by early memtest. If memtest was not run, this field will not 1218 be displayed at all. Size is never rounded down to 0 kB. 1219 That means if 0 kB is reported, you can safely assume 1220 there was at least one pass of memtest and none of the passes 1221 found a single faulty byte of RAM. 1222HardwareCorrupted 1223 The amount of RAM/memory in KB, the kernel identifies as 1224 corrupted. 1225AnonHugePages 1226 Non-file backed huge pages mapped into userspace page tables 1227ShmemHugePages 1228 Memory used by shared memory (shmem) and tmpfs allocated 1229 with huge pages 1230ShmemPmdMapped 1231 Shared memory mapped into userspace with huge pages 1232FileHugePages 1233 Memory used for filesystem data (page cache) allocated 1234 with huge pages 1235FilePmdMapped 1236 Page cache mapped into userspace with huge pages 1237CmaTotal 1238 Memory reserved for the Contiguous Memory Allocator (CMA) 1239CmaFree 1240 Free remaining memory in the CMA reserves 1241HugePages_Total, HugePages_Free, HugePages_Rsvd, HugePages_Surp, Hugepagesize, Hugetlb 1242 See Documentation/admin-guide/mm/hugetlbpage.rst. 1243DirectMap4k, DirectMap2M, DirectMap1G 1244 Breakdown of page table sizes used in the kernel's 1245 identity mapping of RAM 1246 1247vmallocinfo 1248~~~~~~~~~~~ 1249 1250Provides information about vmalloced/vmaped areas. One line per area, 1251containing the virtual address range of the area, size in bytes, 1252caller information of the creator, and optional information depending 1253on the kind of area: 1254 1255 ========== =================================================== 1256 pages=nr number of pages 1257 phys=addr if a physical address was specified 1258 ioremap I/O mapping (ioremap() and friends) 1259 vmalloc vmalloc() area 1260 vmap vmap()ed pages 1261 user VM_USERMAP area 1262 vpages buffer for pages pointers was vmalloced (huge area) 1263 N<node>=nr (Only on NUMA kernels) 1264 Number of pages allocated on memory node <node> 1265 ========== =================================================== 1266 1267:: 1268 1269 > cat /proc/vmallocinfo 1270 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 1271 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 1272 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 1273 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 1274 0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 1275 phys=7fee8000 ioremap 1276 0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 1277 phys=7fee7000 ioremap 1278 0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 1279 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 1280 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 1281 0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 1282 pages=2 vmalloc N1=2 1283 0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 1284 /0x130 [x_tables] pages=4 vmalloc N0=4 1285 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 1286 pages=14 vmalloc N2=14 1287 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 1288 pages=4 vmalloc N1=4 1289 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 1290 pages=2 vmalloc N1=2 1291 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1292 pages=10 vmalloc N0=10 1293 1294 1295softirqs 1296~~~~~~~~ 1297 1298Provides counts of softirq handlers serviced since boot time, for each CPU. 1299 1300:: 1301 1302 > cat /proc/softirqs 1303 CPU0 CPU1 CPU2 CPU3 1304 HI: 0 0 0 0 1305 TIMER: 27166 27120 27097 27034 1306 NET_TX: 0 0 0 17 1307 NET_RX: 42 0 0 39 1308 BLOCK: 0 0 107 1121 1309 TASKLET: 0 0 0 290 1310 SCHED: 27035 26983 26971 26746 1311 HRTIMER: 0 0 0 0 1312 RCU: 1678 1769 2178 2250 1313 13141.3 Networking info in /proc/net 1315-------------------------------- 1316 1317The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 1318additional values you get for IP version 6 if you configure the kernel to 1319support this. Table 1-9 lists the files and their meaning. 1320 1321 1322.. table:: Table 1-8: IPv6 info in /proc/net 1323 1324 ========== ===================================================== 1325 File Content 1326 ========== ===================================================== 1327 udp6 UDP sockets (IPv6) 1328 tcp6 TCP sockets (IPv6) 1329 raw6 Raw device statistics (IPv6) 1330 igmp6 IP multicast addresses, which this host joined (IPv6) 1331 if_inet6 List of IPv6 interface addresses 1332 ipv6_route Kernel routing table for IPv6 1333 rt6_stats Global IPv6 routing tables statistics 1334 sockstat6 Socket statistics (IPv6) 1335 snmp6 Snmp data (IPv6) 1336 ========== ===================================================== 1337 1338.. table:: Table 1-9: Network info in /proc/net 1339 1340 ============= ================================================================ 1341 File Content 1342 ============= ================================================================ 1343 arp Kernel ARP table 1344 dev network devices with statistics 1345 dev_mcast the Layer2 multicast groups a device is listening too 1346 (interface index, label, number of references, number of bound 1347 addresses). 1348 dev_stat network device status 1349 ip_fwchains Firewall chain linkage 1350 ip_fwnames Firewall chain names 1351 ip_masq Directory containing the masquerading tables 1352 ip_masquerade Major masquerading table 1353 netstat Network statistics 1354 raw raw device statistics 1355 route Kernel routing table 1356 rpc Directory containing rpc info 1357 rt_cache Routing cache 1358 snmp SNMP data 1359 sockstat Socket statistics 1360 softnet_stat Per-CPU incoming packets queues statistics of online CPUs 1361 tcp TCP sockets 1362 udp UDP sockets 1363 unix UNIX domain sockets 1364 wireless Wireless interface data (Wavelan etc) 1365 igmp IP multicast addresses, which this host joined 1366 psched Global packet scheduler parameters. 1367 netlink List of PF_NETLINK sockets 1368 ip_mr_vifs List of multicast virtual interfaces 1369 ip_mr_cache List of multicast routing cache 1370 ============= ================================================================ 1371 1372You can use this information to see which network devices are available in 1373your system and how much traffic was routed over those devices:: 1374 1375 > cat /proc/net/dev 1376 Inter-|Receive |[... 1377 face |bytes packets errs drop fifo frame compressed multicast|[... 1378 lo: 908188 5596 0 0 0 0 0 0 [... 1379 ppp0:15475140 20721 410 0 0 410 0 0 [... 1380 eth0: 614530 7085 0 0 0 0 0 1 [... 1381 1382 ...] Transmit 1383 ...] bytes packets errs drop fifo colls carrier compressed 1384 ...] 908188 5596 0 0 0 0 0 0 1385 ...] 1375103 17405 0 0 0 0 0 0 1386 ...] 1703981 5535 0 0 0 3 0 0 1387 1388In addition, each Channel Bond interface has its own directory. For 1389example, the bond0 device will have a directory called /proc/net/bond0/. 1390It will contain information that is specific to that bond, such as the 1391current slaves of the bond, the link status of the slaves, and how 1392many times the slaves link has failed. 1393 13941.4 SCSI info 1395------------- 1396 1397If you have a SCSI or ATA host adapter in your system, you'll find a 1398subdirectory named after the driver for this adapter in /proc/scsi. 1399You'll also see a list of all recognized SCSI devices in /proc/scsi:: 1400 1401 >cat /proc/scsi/scsi 1402 Attached devices: 1403 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1404 Vendor: IBM Model: DGHS09U Rev: 03E0 1405 Type: Direct-Access ANSI SCSI revision: 03 1406 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1407 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1408 Type: CD-ROM ANSI SCSI revision: 02 1409 1410 1411The directory named after the driver has one file for each adapter found in 1412the system. These files contain information about the controller, including 1413the used IRQ and the IO address range. The amount of information shown is 1414dependent on the adapter you use. The example shows the output for an Adaptec 1415AHA-2940 SCSI adapter:: 1416 1417 > cat /proc/scsi/aic7xxx/0 1418 1419 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1420 Compile Options: 1421 TCQ Enabled By Default : Disabled 1422 AIC7XXX_PROC_STATS : Disabled 1423 AIC7XXX_RESET_DELAY : 5 1424 Adapter Configuration: 1425 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1426 Ultra Wide Controller 1427 PCI MMAPed I/O Base: 0xeb001000 1428 Adapter SEEPROM Config: SEEPROM found and used. 1429 Adaptec SCSI BIOS: Enabled 1430 IRQ: 10 1431 SCBs: Active 0, Max Active 2, 1432 Allocated 15, HW 16, Page 255 1433 Interrupts: 160328 1434 BIOS Control Word: 0x18b6 1435 Adapter Control Word: 0x005b 1436 Extended Translation: Enabled 1437 Disconnect Enable Flags: 0xffff 1438 Ultra Enable Flags: 0x0001 1439 Tag Queue Enable Flags: 0x0000 1440 Ordered Queue Tag Flags: 0x0000 1441 Default Tag Queue Depth: 8 1442 Tagged Queue By Device array for aic7xxx host instance 0: 1443 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1444 Actual queue depth per device for aic7xxx host instance 0: 1445 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1446 Statistics: 1447 (scsi0:0:0:0) 1448 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1449 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1450 Total transfers 160151 (74577 reads and 85574 writes) 1451 (scsi0:0:6:0) 1452 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1453 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1454 Total transfers 0 (0 reads and 0 writes) 1455 1456 14571.5 Parallel port info in /proc/parport 1458--------------------------------------- 1459 1460The directory /proc/parport contains information about the parallel ports of 1461your system. It has one subdirectory for each port, named after the port 1462number (0,1,2,...). 1463 1464These directories contain the four files shown in Table 1-10. 1465 1466 1467.. table:: Table 1-10: Files in /proc/parport 1468 1469 ========= ==================================================================== 1470 File Content 1471 ========= ==================================================================== 1472 autoprobe Any IEEE-1284 device ID information that has been acquired. 1473 devices list of the device drivers using that port. A + will appear by the 1474 name of the device currently using the port (it might not appear 1475 against any). 1476 hardware Parallel port's base address, IRQ line and DMA channel. 1477 irq IRQ that parport is using for that port. This is in a separate 1478 file to allow you to alter it by writing a new value in (IRQ 1479 number or none). 1480 ========= ==================================================================== 1481 14821.6 TTY info in /proc/tty 1483------------------------- 1484 1485Information about the available and actually used tty's can be found in the 1486directory /proc/tty. You'll find entries for drivers and line disciplines in 1487this directory, as shown in Table 1-11. 1488 1489 1490.. table:: Table 1-11: Files in /proc/tty 1491 1492 ============= ============================================== 1493 File Content 1494 ============= ============================================== 1495 drivers list of drivers and their usage 1496 ldiscs registered line disciplines 1497 driver/serial usage statistic and status of single tty lines 1498 ============= ============================================== 1499 1500To see which tty's are currently in use, you can simply look into the file 1501/proc/tty/drivers:: 1502 1503 > cat /proc/tty/drivers 1504 pty_slave /dev/pts 136 0-255 pty:slave 1505 pty_master /dev/ptm 128 0-255 pty:master 1506 pty_slave /dev/ttyp 3 0-255 pty:slave 1507 pty_master /dev/pty 2 0-255 pty:master 1508 serial /dev/cua 5 64-67 serial:callout 1509 serial /dev/ttyS 4 64-67 serial 1510 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1511 /dev/ptmx /dev/ptmx 5 2 system 1512 /dev/console /dev/console 5 1 system:console 1513 /dev/tty /dev/tty 5 0 system:/dev/tty 1514 unknown /dev/tty 4 1-63 console 1515 1516 15171.7 Miscellaneous kernel statistics in /proc/stat 1518------------------------------------------------- 1519 1520Various pieces of information about kernel activity are available in the 1521/proc/stat file. All of the numbers reported in this file are aggregates 1522since the system first booted. For a quick look, simply cat the file:: 1523 1524 > cat /proc/stat 1525 cpu 237902850 368826709 106375398 1873517540 1135548 0 14507935 0 0 0 1526 cpu0 60045249 91891769 26331539 468411416 495718 0 5739640 0 0 0 1527 cpu1 59746288 91759249 26609887 468860630 312281 0 4384817 0 0 0 1528 cpu2 59489247 92985423 26904446 467808813 171668 0 2268998 0 0 0 1529 cpu3 58622065 92190267 26529524 468436680 155879 0 2114478 0 0 0 1530 intr 8688370575 8 3373 0 0 0 0 0 0 1 40791 0 0 353317 0 0 0 0 224789828 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 190974333 41958554 123983334 43 0 224593 0 0 0 <more 0's deleted> 1531 ctxt 22848221062 1532 btime 1605316999 1533 processes 746787147 1534 procs_running 2 1535 procs_blocked 0 1536 softirq 12121874454 100099120 3938138295 127375644 2795979 187870761 0 173808342 3072582055 52608 224184354 1537 1538The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1539lines. These numbers identify the amount of time the CPU has spent performing 1540different kinds of work. Time units are in USER_HZ (typically hundredths of a 1541second). The meanings of the columns are as follows, from left to right: 1542 1543- user: normal processes executing in user mode 1544- nice: niced processes executing in user mode 1545- system: processes executing in kernel mode 1546- idle: twiddling thumbs 1547- iowait: In a word, iowait stands for waiting for I/O to complete. But there 1548 are several problems: 1549 1550 1. CPU will not wait for I/O to complete, iowait is the time that a task is 1551 waiting for I/O to complete. When CPU goes into idle state for 1552 outstanding task I/O, another task will be scheduled on this CPU. 1553 2. In a multi-core CPU, the task waiting for I/O to complete is not running 1554 on any CPU, so the iowait of each CPU is difficult to calculate. 1555 3. The value of iowait field in /proc/stat will decrease in certain 1556 conditions. 1557 1558 So, the iowait is not reliable by reading from /proc/stat. 1559- irq: servicing interrupts 1560- softirq: servicing softirqs 1561- steal: involuntary wait 1562- guest: running a normal guest 1563- guest_nice: running a niced guest 1564 1565The "intr" line gives counts of interrupts serviced since boot time, for each 1566of the possible system interrupts. The first column is the total of all 1567interrupts serviced including unnumbered architecture specific interrupts; 1568each subsequent column is the total for that particular numbered interrupt. 1569Unnumbered interrupts are not shown, only summed into the total. 1570 1571The "ctxt" line gives the total number of context switches across all CPUs. 1572 1573The "btime" line gives the time at which the system booted, in seconds since 1574the Unix epoch. 1575 1576The "processes" line gives the number of processes and threads created, which 1577includes (but is not limited to) those created by calls to the fork() and 1578clone() system calls. 1579 1580The "procs_running" line gives the total number of threads that are 1581running or ready to run (i.e., the total number of runnable threads). 1582 1583The "procs_blocked" line gives the number of processes currently blocked, 1584waiting for I/O to complete. 1585 1586The "softirq" line gives counts of softirqs serviced since boot time, for each 1587of the possible system softirqs. The first column is the total of all 1588softirqs serviced; each subsequent column is the total for that particular 1589softirq. 1590 1591 15921.8 Ext4 file system parameters 1593------------------------------- 1594 1595Information about mounted ext4 file systems can be found in 1596/proc/fs/ext4. Each mounted filesystem will have a directory in 1597/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1598/proc/fs/ext4/sda9 or /proc/fs/ext4/dm-0). The files in each per-device 1599directory are shown in Table 1-12, below. 1600 1601.. table:: Table 1-12: Files in /proc/fs/ext4/<devname> 1602 1603 ============== ========================================================== 1604 File Content 1605 mb_groups details of multiblock allocator buddy cache of free blocks 1606 ============== ========================================================== 1607 16081.9 /proc/consoles 1609------------------- 1610Shows registered system console lines. 1611 1612To see which character device lines are currently used for the system console 1613/dev/console, you may simply look into the file /proc/consoles:: 1614 1615 > cat /proc/consoles 1616 tty0 -WU (ECp) 4:7 1617 ttyS0 -W- (Ep) 4:64 1618 1619The columns are: 1620 1621+--------------------+-------------------------------------------------------+ 1622| device | name of the device | 1623+====================+=======================================================+ 1624| operations | * R = can do read operations | 1625| | * W = can do write operations | 1626| | * U = can do unblank | 1627+--------------------+-------------------------------------------------------+ 1628| flags | * E = it is enabled | 1629| | * C = it is preferred console | 1630| | * B = it is primary boot console | 1631| | * p = it is used for printk buffer | 1632| | * b = it is not a TTY but a Braille device | 1633| | * a = it is safe to use when cpu is offline | 1634+--------------------+-------------------------------------------------------+ 1635| major:minor | major and minor number of the device separated by a | 1636| | colon | 1637+--------------------+-------------------------------------------------------+ 1638 1639Summary 1640------- 1641 1642The /proc file system serves information about the running system. It not only 1643allows access to process data but also allows you to request the kernel status 1644by reading files in the hierarchy. 1645 1646The directory structure of /proc reflects the types of information and makes 1647it easy, if not obvious, where to look for specific data. 1648 1649Chapter 2: Modifying System Parameters 1650====================================== 1651 1652In This Chapter 1653--------------- 1654 1655* Modifying kernel parameters by writing into files found in /proc/sys 1656* Exploring the files which modify certain parameters 1657* Review of the /proc/sys file tree 1658 1659------------------------------------------------------------------------------ 1660 1661A very interesting part of /proc is the directory /proc/sys. This is not only 1662a source of information, it also allows you to change parameters within the 1663kernel. Be very careful when attempting this. You can optimize your system, 1664but you can also cause it to crash. Never alter kernel parameters on a 1665production system. Set up a development machine and test to make sure that 1666everything works the way you want it to. You may have no alternative but to 1667reboot the machine once an error has been made. 1668 1669To change a value, simply echo the new value into the file. 1670You need to be root to do this. You can create your own boot script 1671to perform this every time your system boots. 1672 1673The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1674general things in the operation of the Linux kernel. Since some of the files 1675can inadvertently disrupt your system, it is advisable to read both 1676documentation and source before actually making adjustments. In any case, be 1677very careful when writing to any of these files. The entries in /proc may 1678change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1679review the kernel documentation in the directory linux/Documentation. 1680This chapter is heavily based on the documentation included in the pre 2.2 1681kernels, and became part of it in version 2.2.1 of the Linux kernel. 1682 1683Please see: Documentation/admin-guide/sysctl/ directory for descriptions of 1684these entries. 1685 1686Summary 1687------- 1688 1689Certain aspects of kernel behavior can be modified at runtime, without the 1690need to recompile the kernel, or even to reboot the system. The files in the 1691/proc/sys tree can not only be read, but also modified. You can use the echo 1692command to write value into these files, thereby changing the default settings 1693of the kernel. 1694 1695 1696Chapter 3: Per-process Parameters 1697================================= 1698 16993.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1700-------------------------------------------------------------------------------- 1701 1702These files can be used to adjust the badness heuristic used to select which 1703process gets killed in out of memory (oom) conditions. 1704 1705The badness heuristic assigns a value to each candidate task ranging from 0 1706(never kill) to 1000 (always kill) to determine which process is targeted. The 1707units are roughly a proportion along that range of allowed memory the process 1708may allocate from based on an estimation of its current memory and swap use. 1709For example, if a task is using all allowed memory, its badness score will be 17101000. If it is using half of its allowed memory, its score will be 500. 1711 1712The amount of "allowed" memory depends on the context in which the oom killer 1713was called. If it is due to the memory assigned to the allocating task's cpuset 1714being exhausted, the allowed memory represents the set of mems assigned to that 1715cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1716memory represents the set of mempolicy nodes. If it is due to a memory 1717limit (or swap limit) being reached, the allowed memory is that configured 1718limit. Finally, if it is due to the entire system being out of memory, the 1719allowed memory represents all allocatable resources. 1720 1721The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1722is used to determine which task to kill. Acceptable values range from -1000 1723(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1724polarize the preference for oom killing either by always preferring a certain 1725task or completely disabling it. The lowest possible value, -1000, is 1726equivalent to disabling oom killing entirely for that task since it will always 1727report a badness score of 0. 1728 1729Consequently, it is very simple for userspace to define the amount of memory to 1730consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1731example, is roughly equivalent to allowing the remainder of tasks sharing the 1732same system, cpuset, mempolicy, or memory controller resources to use at least 173350% more memory. A value of -500, on the other hand, would be roughly 1734equivalent to discounting 50% of the task's allowed memory from being considered 1735as scoring against the task. 1736 1737For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1738be used to tune the badness score. Its acceptable values range from -16 1739(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1740(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1741scaled linearly with /proc/<pid>/oom_score_adj. 1742 1743The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1744value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1745requires CAP_SYS_RESOURCE. 1746 1747 17483.2 /proc/<pid>/oom_score - Display current oom-killer score 1749------------------------------------------------------------- 1750 1751This file can be used to check the current score used by the oom-killer for 1752any given <pid>. Use it together with /proc/<pid>/oom_score_adj to tune which 1753process should be killed in an out-of-memory situation. 1754 1755Please note that the exported value includes oom_score_adj so it is 1756effectively in range [0,2000]. 1757 1758 17593.3 /proc/<pid>/io - Display the IO accounting fields 1760------------------------------------------------------- 1761 1762This file contains IO statistics for each running process. 1763 1764Example 1765~~~~~~~ 1766 1767:: 1768 1769 test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1770 [1] 3828 1771 1772 test:/tmp # cat /proc/3828/io 1773 rchar: 323934931 1774 wchar: 323929600 1775 syscr: 632687 1776 syscw: 632675 1777 read_bytes: 0 1778 write_bytes: 323932160 1779 cancelled_write_bytes: 0 1780 1781 1782Description 1783~~~~~~~~~~~ 1784 1785rchar 1786^^^^^ 1787 1788I/O counter: chars read 1789The number of bytes which this task has caused to be read from storage. This 1790is simply the sum of bytes which this process passed to read() and pread(). 1791It includes things like tty IO and it is unaffected by whether or not actual 1792physical disk IO was required (the read might have been satisfied from 1793pagecache). 1794 1795 1796wchar 1797^^^^^ 1798 1799I/O counter: chars written 1800The number of bytes which this task has caused, or shall cause to be written 1801to disk. Similar caveats apply here as with rchar. 1802 1803 1804syscr 1805^^^^^ 1806 1807I/O counter: read syscalls 1808Attempt to count the number of read I/O operations, i.e. syscalls like read() 1809and pread(). 1810 1811 1812syscw 1813^^^^^ 1814 1815I/O counter: write syscalls 1816Attempt to count the number of write I/O operations, i.e. syscalls like 1817write() and pwrite(). 1818 1819 1820read_bytes 1821^^^^^^^^^^ 1822 1823I/O counter: bytes read 1824Attempt to count the number of bytes which this process really did cause to 1825be fetched from the storage layer. Done at the submit_bio() level, so it is 1826accurate for block-backed filesystems. <please add status regarding NFS and 1827CIFS at a later time> 1828 1829 1830write_bytes 1831^^^^^^^^^^^ 1832 1833I/O counter: bytes written 1834Attempt to count the number of bytes which this process caused to be sent to 1835the storage layer. This is done at page-dirtying time. 1836 1837 1838cancelled_write_bytes 1839^^^^^^^^^^^^^^^^^^^^^ 1840 1841The big inaccuracy here is truncate. If a process writes 1MB to a file and 1842then deletes the file, it will in fact perform no writeout. But it will have 1843been accounted as having caused 1MB of write. 1844In other words: The number of bytes which this process caused to not happen, 1845by truncating pagecache. A task can cause "negative" IO too. If this task 1846truncates some dirty pagecache, some IO which another task has been accounted 1847for (in its write_bytes) will not be happening. We _could_ just subtract that 1848from the truncating task's write_bytes, but there is information loss in doing 1849that. 1850 1851 1852.. Note:: 1853 1854 At its current implementation state, this is a bit racy on 32-bit machines: 1855 if process A reads process B's /proc/pid/io while process B is updating one 1856 of those 64-bit counters, process A could see an intermediate result. 1857 1858 1859More information about this can be found within the taskstats documentation in 1860Documentation/accounting. 1861 18623.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1863--------------------------------------------------------------- 1864When a process is dumped, all anonymous memory is written to a core file as 1865long as the size of the core file isn't limited. But sometimes we don't want 1866to dump some memory segments, for example, huge shared memory or DAX. 1867Conversely, sometimes we want to save file-backed memory segments into a core 1868file, not only the individual files. 1869 1870/proc/<pid>/coredump_filter allows you to customize which memory segments 1871will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1872of memory types. If a bit of the bitmask is set, memory segments of the 1873corresponding memory type are dumped, otherwise they are not dumped. 1874 1875The following 9 memory types are supported: 1876 1877 - (bit 0) anonymous private memory 1878 - (bit 1) anonymous shared memory 1879 - (bit 2) file-backed private memory 1880 - (bit 3) file-backed shared memory 1881 - (bit 4) ELF header pages in file-backed private memory areas (it is 1882 effective only if the bit 2 is cleared) 1883 - (bit 5) hugetlb private memory 1884 - (bit 6) hugetlb shared memory 1885 - (bit 7) DAX private memory 1886 - (bit 8) DAX shared memory 1887 1888 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1889 are always dumped regardless of the bitmask status. 1890 1891 Note that bits 0-4 don't affect hugetlb or DAX memory. hugetlb memory is 1892 only affected by bit 5-6, and DAX is only affected by bits 7-8. 1893 1894The default value of coredump_filter is 0x33; this means all anonymous memory 1895segments, ELF header pages and hugetlb private memory are dumped. 1896 1897If you don't want to dump all shared memory segments attached to pid 1234, 1898write 0x31 to the process's proc file:: 1899 1900 $ echo 0x31 > /proc/1234/coredump_filter 1901 1902When a new process is created, the process inherits the bitmask status from its 1903parent. It is useful to set up coredump_filter before the program runs. 1904For example:: 1905 1906 $ echo 0x7 > /proc/self/coredump_filter 1907 $ ./some_program 1908 19093.5 /proc/<pid>/mountinfo - Information about mounts 1910-------------------------------------------------------- 1911 1912This file contains lines of the form:: 1913 1914 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1915 (1)(2)(3) (4) (5) (6) (n…m) (m+1)(m+2) (m+3) (m+4) 1916 1917 (1) mount ID: unique identifier of the mount (may be reused after umount) 1918 (2) parent ID: ID of parent (or of self for the top of the mount tree) 1919 (3) major:minor: value of st_dev for files on filesystem 1920 (4) root: root of the mount within the filesystem 1921 (5) mount point: mount point relative to the process's root 1922 (6) mount options: per mount options 1923 (n…m) optional fields: zero or more fields of the form "tag[:value]" 1924 (m+1) separator: marks the end of the optional fields 1925 (m+2) filesystem type: name of filesystem of the form "type[.subtype]" 1926 (m+3) mount source: filesystem specific information or "none" 1927 (m+4) super options: per super block options 1928 1929Parsers should ignore all unrecognised optional fields. Currently the 1930possible optional fields are: 1931 1932================ ============================================================== 1933shared:X mount is shared in peer group X 1934master:X mount is slave to peer group X 1935propagate_from:X mount is slave and receives propagation from peer group X [#]_ 1936unbindable mount is unbindable 1937================ ============================================================== 1938 1939.. [#] X is the closest dominant peer group under the process's root. If 1940 X is the immediate master of the mount, or if there's no dominant peer 1941 group under the same root, then only the "master:X" field is present 1942 and not the "propagate_from:X" field. 1943 1944For more information on mount propagation see: 1945 1946 Documentation/filesystems/sharedsubtree.rst 1947 1948 19493.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1950-------------------------------------------------------- 1951These files provide a method to access a task's comm value. It also allows for 1952a task to set its own or one of its thread siblings comm value. The comm value 1953is limited in size compared to the cmdline value, so writing anything longer 1954then the kernel's TASK_COMM_LEN (currently 16 chars, including the NUL 1955terminator) will result in a truncated comm value. 1956 1957 19583.7 /proc/<pid>/task/<tid>/children - Information about task children 1959------------------------------------------------------------------------- 1960This file provides a fast way to retrieve first level children pids 1961of a task pointed by <pid>/<tid> pair. The format is a space separated 1962stream of pids. 1963 1964Note the "first level" here -- if a child has its own children they will 1965not be listed here; one needs to read /proc/<children-pid>/task/<tid>/children 1966to obtain the descendants. 1967 1968Since this interface is intended to be fast and cheap it doesn't 1969guarantee to provide precise results and some children might be 1970skipped, especially if they've exited right after we printed their 1971pids, so one needs to either stop or freeze processes being inspected 1972if precise results are needed. 1973 1974 19753.8 /proc/<pid>/fdinfo/<fd> - Information about opened file 1976--------------------------------------------------------------- 1977This file provides information associated with an opened file. The regular 1978files have at least four fields -- 'pos', 'flags', 'mnt_id' and 'ino'. 1979The 'pos' represents the current offset of the opened file in decimal 1980form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the 1981file has been created with [see open(2) for details] and 'mnt_id' represents 1982mount ID of the file system containing the opened file [see 3.5 1983/proc/<pid>/mountinfo for details]. 'ino' represents the inode number of 1984the file. 1985 1986A typical output is:: 1987 1988 pos: 0 1989 flags: 0100002 1990 mnt_id: 19 1991 ino: 63107 1992 1993All locks associated with a file descriptor are shown in its fdinfo too:: 1994 1995 lock: 1: FLOCK ADVISORY WRITE 359 00:13:11691 0 EOF 1996 1997The files such as eventfd, fsnotify, signalfd, epoll among the regular pos/flags 1998pair provide additional information particular to the objects they represent. 1999 2000Eventfd files 2001~~~~~~~~~~~~~ 2002 2003:: 2004 2005 pos: 0 2006 flags: 04002 2007 mnt_id: 9 2008 ino: 63107 2009 eventfd-count: 5a 2010 2011where 'eventfd-count' is hex value of a counter. 2012 2013Signalfd files 2014~~~~~~~~~~~~~~ 2015 2016:: 2017 2018 pos: 0 2019 flags: 04002 2020 mnt_id: 9 2021 ino: 63107 2022 sigmask: 0000000000000200 2023 2024where 'sigmask' is hex value of the signal mask associated 2025with a file. 2026 2027Epoll files 2028~~~~~~~~~~~ 2029 2030:: 2031 2032 pos: 0 2033 flags: 02 2034 mnt_id: 9 2035 ino: 63107 2036 tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7 2037 2038where 'tfd' is a target file descriptor number in decimal form, 2039'events' is events mask being watched and the 'data' is data 2040associated with a target [see epoll(7) for more details]. 2041 2042The 'pos' is current offset of the target file in decimal form 2043[see lseek(2)], 'ino' and 'sdev' are inode and device numbers 2044where target file resides, all in hex format. 2045 2046Fsnotify files 2047~~~~~~~~~~~~~~ 2048For inotify files the format is the following:: 2049 2050 pos: 0 2051 flags: 02000000 2052 mnt_id: 9 2053 ino: 63107 2054 inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d 2055 2056where 'wd' is a watch descriptor in decimal form, i.e. a target file 2057descriptor number, 'ino' and 'sdev' are inode and device where the 2058target file resides and the 'mask' is the mask of events, all in hex 2059form [see inotify(7) for more details]. 2060 2061If the kernel was built with exportfs support, the path to the target 2062file is encoded as a file handle. The file handle is provided by three 2063fields 'fhandle-bytes', 'fhandle-type' and 'f_handle', all in hex 2064format. 2065 2066If the kernel is built without exportfs support the file handle won't be 2067printed out. 2068 2069If there is no inotify mark attached yet the 'inotify' line will be omitted. 2070 2071For fanotify files the format is:: 2072 2073 pos: 0 2074 flags: 02 2075 mnt_id: 9 2076 ino: 63107 2077 fanotify flags:10 event-flags:0 2078 fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003 2079 fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4 2080 2081where fanotify 'flags' and 'event-flags' are values used in fanotify_init 2082call, 'mnt_id' is the mount point identifier, 'mflags' is the value of 2083flags associated with mark which are tracked separately from events 2084mask. 'ino' and 'sdev' are target inode and device, 'mask' is the events 2085mask and 'ignored_mask' is the mask of events which are to be ignored. 2086All are in hex format. Incorporation of 'mflags', 'mask' and 'ignored_mask' 2087provide information about flags and mask used in fanotify_mark 2088call [see fsnotify manpage for details]. 2089 2090While the first three lines are mandatory and always printed, the rest is 2091optional and may be omitted if no marks created yet. 2092 2093Timerfd files 2094~~~~~~~~~~~~~ 2095 2096:: 2097 2098 pos: 0 2099 flags: 02 2100 mnt_id: 9 2101 ino: 63107 2102 clockid: 0 2103 ticks: 0 2104 settime flags: 01 2105 it_value: (0, 49406829) 2106 it_interval: (1, 0) 2107 2108where 'clockid' is the clock type and 'ticks' is the number of the timer expirations 2109that have occurred [see timerfd_create(2) for details]. 'settime flags' are 2110flags in octal form been used to setup the timer [see timerfd_settime(2) for 2111details]. 'it_value' is remaining time until the timer expiration. 2112'it_interval' is the interval for the timer. Note the timer might be set up 2113with TIMER_ABSTIME option which will be shown in 'settime flags', but 'it_value' 2114still exhibits timer's remaining time. 2115 2116DMA Buffer files 2117~~~~~~~~~~~~~~~~ 2118 2119:: 2120 2121 pos: 0 2122 flags: 04002 2123 mnt_id: 9 2124 ino: 63107 2125 size: 32768 2126 count: 2 2127 exp_name: system-heap 2128 2129where 'size' is the size of the DMA buffer in bytes. 'count' is the file count of 2130the DMA buffer file. 'exp_name' is the name of the DMA buffer exporter. 2131 21323.9 /proc/<pid>/map_files - Information about memory mapped files 2133--------------------------------------------------------------------- 2134This directory contains symbolic links which represent memory mapped files 2135the process is maintaining. Example output:: 2136 2137 | lr-------- 1 root root 64 Jan 27 11:24 333c600000-333c620000 -> /usr/lib64/ld-2.18.so 2138 | lr-------- 1 root root 64 Jan 27 11:24 333c81f000-333c820000 -> /usr/lib64/ld-2.18.so 2139 | lr-------- 1 root root 64 Jan 27 11:24 333c820000-333c821000 -> /usr/lib64/ld-2.18.so 2140 | ... 2141 | lr-------- 1 root root 64 Jan 27 11:24 35d0421000-35d0422000 -> /usr/lib64/libselinux.so.1 2142 | lr-------- 1 root root 64 Jan 27 11:24 400000-41a000 -> /usr/bin/ls 2143 2144The name of a link represents the virtual memory bounds of a mapping, i.e. 2145vm_area_struct::vm_start-vm_area_struct::vm_end. 2146 2147The main purpose of the map_files is to retrieve a set of memory mapped 2148files in a fast way instead of parsing /proc/<pid>/maps or 2149/proc/<pid>/smaps, both of which contain many more records. At the same 2150time one can open(2) mappings from the listings of two processes and 2151comparing their inode numbers to figure out which anonymous memory areas 2152are actually shared. 2153 21543.10 /proc/<pid>/timerslack_ns - Task timerslack value 2155--------------------------------------------------------- 2156This file provides the value of the task's timerslack value in nanoseconds. 2157This value specifies an amount of time that normal timers may be deferred 2158in order to coalesce timers and avoid unnecessary wakeups. 2159 2160This allows a task's interactivity vs power consumption tradeoff to be 2161adjusted. 2162 2163Writing 0 to the file will set the task's timerslack to the default value. 2164 2165Valid values are from 0 - ULLONG_MAX 2166 2167An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level 2168permissions on the task specified to change its timerslack_ns value. 2169 21703.11 /proc/<pid>/patch_state - Livepatch patch operation state 2171----------------------------------------------------------------- 2172When CONFIG_LIVEPATCH is enabled, this file displays the value of the 2173patch state for the task. 2174 2175A value of '-1' indicates that no patch is in transition. 2176 2177A value of '0' indicates that a patch is in transition and the task is 2178unpatched. If the patch is being enabled, then the task hasn't been 2179patched yet. If the patch is being disabled, then the task has already 2180been unpatched. 2181 2182A value of '1' indicates that a patch is in transition and the task is 2183patched. If the patch is being enabled, then the task has already been 2184patched. If the patch is being disabled, then the task hasn't been 2185unpatched yet. 2186 21873.12 /proc/<pid>/arch_status - task architecture specific status 2188------------------------------------------------------------------- 2189When CONFIG_PROC_PID_ARCH_STATUS is enabled, this file displays the 2190architecture specific status of the task. 2191 2192Example 2193~~~~~~~ 2194 2195:: 2196 2197 $ cat /proc/6753/arch_status 2198 AVX512_elapsed_ms: 8 2199 2200Description 2201~~~~~~~~~~~ 2202 2203x86 specific entries 2204~~~~~~~~~~~~~~~~~~~~~ 2205 2206AVX512_elapsed_ms 2207^^^^^^^^^^^^^^^^^^ 2208 2209 If AVX512 is supported on the machine, this entry shows the milliseconds 2210 elapsed since the last time AVX512 usage was recorded. The recording 2211 happens on a best effort basis when a task is scheduled out. This means 2212 that the value depends on two factors: 2213 2214 1) The time which the task spent on the CPU without being scheduled 2215 out. With CPU isolation and a single runnable task this can take 2216 several seconds. 2217 2218 2) The time since the task was scheduled out last. Depending on the 2219 reason for being scheduled out (time slice exhausted, syscall ...) 2220 this can be arbitrary long time. 2221 2222 As a consequence the value cannot be considered precise and authoritative 2223 information. The application which uses this information has to be aware 2224 of the overall scenario on the system in order to determine whether a 2225 task is a real AVX512 user or not. Precise information can be obtained 2226 with performance counters. 2227 2228 A special value of '-1' indicates that no AVX512 usage was recorded, thus 2229 the task is unlikely an AVX512 user, but depends on the workload and the 2230 scheduling scenario, it also could be a false negative mentioned above. 2231 22323.13 /proc/<pid>/fd - List of symlinks to open files 2233------------------------------------------------------- 2234This directory contains symbolic links which represent open files 2235the process is maintaining. Example output:: 2236 2237 lr-x------ 1 root root 64 Sep 20 17:53 0 -> /dev/null 2238 l-wx------ 1 root root 64 Sep 20 17:53 1 -> /dev/null 2239 lrwx------ 1 root root 64 Sep 20 17:53 10 -> 'socket:[12539]' 2240 lrwx------ 1 root root 64 Sep 20 17:53 11 -> 'socket:[12540]' 2241 lrwx------ 1 root root 64 Sep 20 17:53 12 -> 'socket:[12542]' 2242 2243The number of open files for the process is stored in 'size' member 2244of stat() output for /proc/<pid>/fd for fast access. 2245------------------------------------------------------- 2246 22473.14 /proc/<pid/ksm_stat - Information about the process's ksm status 2248--------------------------------------------------------------------- 2249When CONFIG_KSM is enabled, each process has this file which displays 2250the information of ksm merging status. 2251 2252Example 2253~~~~~~~ 2254 2255:: 2256 2257 / # cat /proc/self/ksm_stat 2258 ksm_rmap_items 0 2259 ksm_zero_pages 0 2260 ksm_merging_pages 0 2261 ksm_process_profit 0 2262 ksm_merge_any: no 2263 ksm_mergeable: no 2264 2265Description 2266~~~~~~~~~~~ 2267 2268ksm_rmap_items 2269^^^^^^^^^^^^^^ 2270 2271The number of ksm_rmap_item structures in use. The structure 2272ksm_rmap_item stores the reverse mapping information for virtual 2273addresses. KSM will generate a ksm_rmap_item for each ksm-scanned page of 2274the process. 2275 2276ksm_zero_pages 2277^^^^^^^^^^^^^^ 2278 2279When /sys/kernel/mm/ksm/use_zero_pages is enabled, it represent how many 2280empty pages are merged with kernel zero pages by KSM. 2281 2282ksm_merging_pages 2283^^^^^^^^^^^^^^^^^ 2284 2285It represents how many pages of this process are involved in KSM merging 2286(not including ksm_zero_pages). It is the same with what 2287/proc/<pid>/ksm_merging_pages shows. 2288 2289ksm_process_profit 2290^^^^^^^^^^^^^^^^^^ 2291 2292The profit that KSM brings (Saved bytes). KSM can save memory by merging 2293identical pages, but also can consume additional memory, because it needs 2294to generate a number of rmap_items to save each scanned page's brief rmap 2295information. Some of these pages may be merged, but some may not be abled 2296to be merged after being checked several times, which are unprofitable 2297memory consumed. 2298 2299ksm_merge_any 2300^^^^^^^^^^^^^ 2301 2302It specifies whether the process's 'mm is added by prctl() into the 2303candidate list of KSM or not, and if KSM scanning is fully enabled at 2304process level. 2305 2306ksm_mergeable 2307^^^^^^^^^^^^^ 2308 2309It specifies whether any VMAs of the process''s mms are currently 2310applicable to KSM. 2311 2312More information about KSM can be found in 2313Documentation/admin-guide/mm/ksm.rst. 2314 2315 2316Chapter 4: Configuring procfs 2317============================= 2318 23194.1 Mount options 2320--------------------- 2321 2322The following mount options are supported: 2323 2324 ========= ======================================================== 2325 hidepid= Set /proc/<pid>/ access mode. 2326 gid= Set the group authorized to learn processes information. 2327 subset= Show only the specified subset of procfs. 2328 ========= ======================================================== 2329 2330hidepid=off or hidepid=0 means classic mode - everybody may access all 2331/proc/<pid>/ directories (default). 2332 2333hidepid=noaccess or hidepid=1 means users may not access any /proc/<pid>/ 2334directories but their own. Sensitive files like cmdline, sched*, status are now 2335protected against other users. This makes it impossible to learn whether any 2336user runs specific program (given the program doesn't reveal itself by its 2337behaviour). As an additional bonus, as /proc/<pid>/cmdline is unaccessible for 2338other users, poorly written programs passing sensitive information via program 2339arguments are now protected against local eavesdroppers. 2340 2341hidepid=invisible or hidepid=2 means hidepid=1 plus all /proc/<pid>/ will be 2342fully invisible to other users. It doesn't mean that it hides a fact whether a 2343process with a specific pid value exists (it can be learned by other means, e.g. 2344by "kill -0 $PID"), but it hides process's uid and gid, which may be learned by 2345stat()'ing /proc/<pid>/ otherwise. It greatly complicates an intruder's task of 2346gathering information about running processes, whether some daemon runs with 2347elevated privileges, whether other user runs some sensitive program, whether 2348other users run any program at all, etc. 2349 2350hidepid=ptraceable or hidepid=4 means that procfs should only contain 2351/proc/<pid>/ directories that the caller can ptrace. 2352 2353gid= defines a group authorized to learn processes information otherwise 2354prohibited by hidepid=. If you use some daemon like identd which needs to learn 2355information about processes information, just add identd to this group. 2356 2357subset=pid hides all top level files and directories in the procfs that 2358are not related to tasks. 2359 2360Chapter 5: Filesystem behavior 2361============================== 2362 2363Originally, before the advent of pid namespace, procfs was a global file 2364system. It means that there was only one procfs instance in the system. 2365 2366When pid namespace was added, a separate procfs instance was mounted in 2367each pid namespace. So, procfs mount options are global among all 2368mountpoints within the same namespace:: 2369 2370 # grep ^proc /proc/mounts 2371 proc /proc proc rw,relatime,hidepid=2 0 0 2372 2373 # strace -e mount mount -o hidepid=1 -t proc proc /tmp/proc 2374 mount("proc", "/tmp/proc", "proc", 0, "hidepid=1") = 0 2375 +++ exited with 0 +++ 2376 2377 # grep ^proc /proc/mounts 2378 proc /proc proc rw,relatime,hidepid=2 0 0 2379 proc /tmp/proc proc rw,relatime,hidepid=2 0 0 2380 2381and only after remounting procfs mount options will change at all 2382mountpoints:: 2383 2384 # mount -o remount,hidepid=1 -t proc proc /tmp/proc 2385 2386 # grep ^proc /proc/mounts 2387 proc /proc proc rw,relatime,hidepid=1 0 0 2388 proc /tmp/proc proc rw,relatime,hidepid=1 0 0 2389 2390This behavior is different from the behavior of other filesystems. 2391 2392The new procfs behavior is more like other filesystems. Each procfs mount 2393creates a new procfs instance. Mount options affect own procfs instance. 2394It means that it became possible to have several procfs instances 2395displaying tasks with different filtering options in one pid namespace:: 2396 2397 # mount -o hidepid=invisible -t proc proc /proc 2398 # mount -o hidepid=noaccess -t proc proc /tmp/proc 2399 # grep ^proc /proc/mounts 2400 proc /proc proc rw,relatime,hidepid=invisible 0 0 2401 proc /tmp/proc proc rw,relatime,hidepid=noaccess 0 0 2402